func.func @outfeed_enqueue_tuple(%data_1: tensor<3xi32>, %data_2: tensor<4xf32>) -> () {
// CHECK: [[TOKEN:%.*]] = mhlo.create_token  : !mhlo.token
// CHECK: "mhlo.outfeed"([[VAL_0]], [[VAL_1]], [[TOKEN]]) <{outfeed_config = ""}> : (tensor<3xi32>, tensor<4xf32>, !mhlo.token) -> !mhlo.token
  "tf.OutfeedEnqueueTuple"(%data_1, %data_2) : (tensor<3xi32>, tensor<4xf32>) -> ()
  func.return
}

//
// Assume segment ids' shape is [SI0, SI1, ..., SIm] and data's  shape is
// [D0, D1, ..., Dn]. Note that segment ids' shape must be a prefix of data's
// shape, so we can have data's shape represented as [SI0, SI1, ..., SIm,
// Dm+1, ..., Dn]. Then
//   output[segment_ids[SI_i0, SI_i1, ..., SI_im], D_im+1, ..., D_in] =
//      <ReductionOp> over data[SI_i0, SI_i1, ..., SI_im, D_im+1, ..., D_in]